>scott931 at uidaho.edu (Patrick Scott) wrote:
>
> My question is: with hybrids such as this (Or any bird, I suppose, which
> defies ID by observation), how does one find out for sure
> what species the bird is?
>
On Thu, 16 Jan 1997, Alan Richards added:
>
> I also have been curious about this -- if one has
> the bird in-hand, is there some kind of DNA analysis,
> or what? that happens -- I vaguely remember reading
> about birds wh/ were xx% Glaucous-winged, or some
> such -- is this really measurable, or was this some
> sort of stochastic guesswork?

In answer to Pat's question, while molecular techniques may not always be
necessary or useful, there are some sticky cases in which they have proved
uniquely able to provide a concrete answer to the question "What the *&^(
is that?"

A terrific example of this is the case of Cox's Sandpiper, a very rare
bird in the genus _Calidris_. Cox's was described as a new species in
1982, but despite suspicions that it might be a hybrid, no agreement
existed as to whether or not it was a hybrid, or even who the parents
would have been if it was a hybrid. Nominees for parental species
included Pectoral, Curlew, and Sharp-tailed Sandpipers, and also Ruff.

In the August 1996 issue of the Condor, L. Christidis, K. Davies, M.
Westerman, P. Christian and R. Schodde published an article titled
"Molecular assesment of the taxonomic status of Cox's Sandpiper." They
looked at mitochondrial DNA sequences and allozyme (variable protein)
differences and - it's a hybrid! The parental species are Pectoral and
Curlew Sandpipers, as was originally suggested, and, what's more, the
authors were able to deduce that for all of the three specimens of Cox's
Sandpiper collected for the study, the Curlew Sandpiper was the mother,
and the Pectoral Sandpiper was the father.

As to Alan Richards's question about being able to say that a certain
individual big pink-footed gull is X percent Glaucous winged and X percent
Western, I doubt that was done by screening the DNA, because of the
difficulty and expense of finding suitable genetic markers. Probably
someone had laid out specimens of pure Westerns, pure Glaucous-wings, and
intermediates along the spectrum. If an observer could judge, for several
plumage or structural characters, where along the spectrum the bird in
question fell, then very likely you could make a guess as to what percent
Glaucous-winged genes it had.

I'd point out that laying out specimens and examining characters is no
less valid than looking directly at the DNA: as long as the characters are
inherited it's the same thing, really. And it's a heck of a lot
cheaper.

So with the mystery bird that started this discussion, mist-netting,
photographing and measuring the bird might provide sufficient evidence to
decide what it is. If not, then probably you could get a definitive
answer from looking at DNA and other molecular markers.

If anyone is ambitious and curious enough, look up the two articles below.
While the first is largely molecular and the second largely focussed on
plumage differences, both articles use all available clues, molecular and
structural, to analyze hybrids.

Christidis, L., K. Davies, M. Westerman, P. D. Christian, and R. Schodde.
1996. Molecular assessment of the taxonomic status of Cox's Sandpiper.
Condor 98: 459-463.

Rohwer, S. 1994. Two new hybrid _Dendroica_ warblers and new methodology
for inferring parental species. Auk 111: 441-449.

Chris Hill
Everett, WA
cehill at u.washington.edu